GIZ NLP Agricultural Keyword Spotter
$7,000 USD
Classify audio utterances in Luganda and English from Uganda
476 data scientists enrolled, 139 on the leaderboard
11 September—29 November
Ends in 30 days
Rules clarification - Metadata - @Zindi?
published 8 Oct 2020, 11:22

@zindi, in the rules it states "Your models should not use any of the metadata provided" - what precisely do you mean by this?

Please don't use the length of utterance or any information from the name of the file. Using metadata will not be useful to the client.

i guess it means we cannot directly extract text from audio and use it ....

Good question, @zindi, do you have a stance on this?

Your task is to predict the labels for the test set. You can extract text from audio.